Digital Tools to Help Preserve Endangered Koryak Language in Russia

Researchers at Kamchatka State University and the School of Linguistics at the Higher School of Economics are using digital tools to preserve the Koryak language. They are building so-called language corpora – annotated digital text databases that can serve as the foundation for translators, digital assistants, and educational applications. In effect, this creates a new environment where the language can exist. Scientists are documenting the language across texts, audio, and archival datasets.

Mobile App Already Available

Within the corpus, words are categorized by form, meaning, and context, after which neural networks are applied. However, standard AI models are not well suited for languages with limited digital resources, so researchers are developing systems that can operate with small datasets. Combined with both manual and automated morphological annotation, this approach allows maximum value to be extracted from limited material and enables the creation of corpora suitable for digital product development. This is the focus of the joint work by the two universities.

A mobile app for learning the Koryak language is already available to the public. Upcoming releases include a weather forecast service in Koryak, an online dictionary, educational video content, and a book.